Shale Oil, Tar Sands, and Related Fuel Sources - ACS Publications

14 Differences Among Ozokerites R O B E R T F . M A R S C H N E R and J. C. W I N T E R S

Downloaded by UNIV LAVAL on July 12, 2016 | http://pubs.acs.org Publication Date: September 1, 1976 | doi: 10.1021/ba-1976-0151.ch014

Research and Development Department, Amoco Oil Co., Naperville, Ill. 60540

Ozokerites are mainly mixtures of n-alkanes that occasionally accompany deposits of petroleum, coal, or lignite. Gas chromatograms of 10 ozokerites from Galicia, Utah, and Russia showed systematic differences in composition. In all samples from Galicia, n-alkanes near C were most abundant. This same maximum probably occurs in hatchettites and Uinta Basin petroleum. Some Utah samples also peaked near C , but in others the maximum occurred near C , with a secondary peak around C . The Russian ozokerites, n-alkanes near either C or C are the more abundant. Multiple favored abundance ranges suggest that a sequence of proccesses were involved in the formation of ozokerites. 29

29

46

31

31

46

"T\eposits of mineral wax occasionally associated with petroleum, coal, lignite, and other native organic substances are known by a profusion of names. If found near oil fields, they are called earth waxes or ozokerites; if found near coal beds, they are usually called hatchettites; if found near lignite, they are sometimes called scheererite; and specimens found i n various environments have been called aragotite, evenldte, fichtelite, hartite, idrialite, posepnyite, valaite, etc. One waxy mineral, kabaite, found in traces i n meteorites, seems not to be a member of the same family and may result from hydrogénation of carbon monoxide ( I ) . Otherwise, variations within most of these minerals are about as large as the differences between them, and the names have consequently not been used rigorously. A generic term ozokerites could suffice for all, and it is so used here. Ozokerites are widely distributed, having been found i n perhaps a hundred localities around the world. Some properties of ozokerites from a selection of major deposits are given i n Table I. M a n y of these deposits are i n the mountainous regions of eastern Europe. The densest occur166 Yen; Shale Oil, Tar Sands, and Related Fuel Sources Advances in Chemistry; American Chemical Society: Washington, DC, 1976.

14.

MARSCHNER

AND WINTERS

167

Ozokerites

rence is i n Galicia along the northeast flank of the Eastern Carpathian Mountains, especially near the town of Borislav, now i n the U S S R but, since the turn of the century, also i n Poland and Austria-Hungary. It occurs with a variety of inorganic minerals and varies i n color from yellow to brown and in melting point from about 40° to 80 °C. n-Alkanes are clearly the characteristic components of ozokerites. Densities and refractive indices are too low for anything but paraffinic hydrocarbons, and the high melting points and low solubilities preclude much of anything other than the normal ones. Unquestionably some


Table I. Location

Characteristics of Ozokerites from Major Deposits Area (Country)

Mineral. Association Color

M.P. (°C)

Reference

60-^80

2

77.5

3

59-65

4

60-79

5

93

6

yellow

49

6

shales

green

38-65

7

limestone

yellowish

38-29

8

44-46

6

Associated with Petroleum Borislav Starunya Shor-su Uinta Basin Slanik

Galicia (USSR) Ukraine (USSR) Fergana (USSR) Utah (USA) Moldavia (Romania)

yellow

gypsum

grayish yellow dark

limestone o i l shale shale, c o a l

Associated with Coal Loch Fyne

P o r t e de France

Scotland (UK) South Wales (UK) Grenoble (France)

St. G a l l e n

(Switzerland)

Pencoed

(Ozokerite)

sandstone

—

Associated with Lignite

—

yellowbrown

— (Hatchettite)

(Scheererite) reddish

branched and cyclic structures are also present (7,9), and the branched structures include monomethylalkanes and the same group of C i - C o isoprenoids that occurs i n petroleum (4). The technology of ozokerites followed a meteoric pattern (10). F r o m first production around I860 until W o r l d W a r I, use for candles, crayons, electrical insulation, and leather dressings expanded steadily. Such elementary refining as was developed consisted of little more than melting the raw wax i n hot water to separate inorganic matter and treating it w i t h concentrated sulfuric acid to produce ceresin, which was lighter i n 5

Yen; Shale Oil, Tar Sands, and Related Fuel Sources Advances in Chemistry; American Chemical Society: Washington, DC, 1976.

2

168

SHALE OIL, TAR SANDS, A N D R E L A T E D F U E L SOURCES


color and higher i n price. Uses proliferated between the Wars. A t first, ozokerite was widely used as a substitute and adulterant for costly bees wax and plant waxes. Later, cheaper waxes from the expanding petro leum industry were used to extend and replace ozokerite. B y W o r l d W a r II, most commercial sources of ozokerite were depleted or even exhausted. Meanwhile, however, vacuum distillation for manufacture of lubricating oils had become widespread i n refineries, and what are now marketed as ozokerites and ceresins are actually mixtures of motor-oil waxes from vacuum distillates with microcrystalline waxes from vacuum residues. Although the origin of ozokerites is not known, they may have been derived from petroleum. Nature may have precipitated ozokerites i n a manner parallel to the way that man has precipitated sucker-rod wax i n the oil fields or tank-bottom waxes i n refineries. The main processes in both instances are removal of light fractions that act as solvent, lowered temperature, and enough time for the wax crystals to consolidate from the viscous medium. Such a mechanism would be difficult to extend to coal or lignite, however, and provenance from unusual plant matter has been proposed as an alternative explanation. Microorganisms also might well have had an influence (11). Ozokerites in any case provided a logical material for extending pre vious work on the n-alkanes i n crude oils (12). There the abundance of n-alkanes was found to be a significant parameter i n the geochemistry. Table II. Source Serial No. (Date)

Descriptions of

Original Donor

Baryslav, Poland" E18150 (1937) Boryslaw, Galacia" M4963 (1894) Galacia, Austria" E5015 (1894) Kyune, Utah" E13536 (1910) Fort Worth, U t a h " E13067 (1905) Russia" E4979 (1894) Carbon County, U t a h Emery County, U t a h H-23 (unknown) Soldier Summit, U t a h

Industrial and Agricultural Museum of Warsaw Ward's Natural Science Estab. Standard O i l Co. private collection private collection World's Columbian Exposition 6

6

6

0

Geology Museum Geology Museum private collection University of Utah

From Field Museum of Natural History through Bertrand G. Woodland. * From Colorado School of Mines through A . Sherrill Hougton. β


14.

MARSCHNER

AND WINTERS

169

Ozokerites

In the present study, gas chromatography was again used, with programmed-temperature operation as before, but with longer times at high temperature.


Experimental Native ozokerites are not as available as they once were, but pieces do exist i n museums and collections. Six samples obtained from the F i e l d Museum of Natural History i n Chicago through Bertram G. Woodland served as the nucleus for the present study. Properties of these and sup plementary samples are listed in Table II. Especially valuable i n view of the clouded history of ozokerites was a description attached to the sample originally from the Industrial and Agricultural Museum i n Warsaw: "Earth wax i n unrefined state directly after excavation from the mine. F r o m the Association for Earth W a x and Rock O i l Industry at Boryslaw." Gas chromatograms were obtained on a twin Hewlett-Packard model 5750 Research Chromatograph. The dual columns were 6-m lengths of copper tubing 3 mm i n diameter, packed with 3 % O V - 1 on Chromosorb G.-HP (methylsilicone on calcined diatomaceous earth). Samples were introduced as about 10% solutions i n carbon bisulfide. Detection was by hydrogen-flame ionization, the non-sample contribution from the idle column being subtracted from the total contribution of the active column to provide a sample chromatograph corrected for extraneous ionization. Ozokerites Examined Melting Point (°C)

Streak on Paper

See text, cast block

61.2

orange-brown

82

sample N o . 110

63.5

orange-yellow

85

commercial

62.2

brown-orange

93

native paraffin

60.0

brown

89

73.6

light brown

95

83.5

reddish brown

70

84.0 84.6

light brown light brown orange-brown dark brown

85 79 93 95

Other

Description

cast cylinder Soldier Summit probably Utah Wasatch County, mine sample β

Approx. % τι-Alkan

From Reino E . Kallio, University of Illinois, Urbana, 111.



170

SHALE

OIL, T A R SANDS, A N D R E L A T E D

FUEL

SOURCES

The temperature was proprammed from 100 °C at 10 °C per min to 400°C., where it was held as long as meaningful ionization data were recorded. These severe conditions pressed the equipment to the limits of operability. Several trials were usually necessary before the complete chromatogram could be accurately defined, and all samples were run two to five times to confirm the presence or absence of elusive characteristics. Columns had to be replaced several times, adjustments to the procedure were frequent, and compromises with consistency and uniformity had to be made continually. A typical composite chromatogram of three runs on the ozokerite from Russia is shown i n Figure 1. Time i n minutes from injection of sample is given at the bottom of the figure, and the temperature of the oven i n which the columns are heated is given at the top. A partial chroCOLUMN TEMPERATURE, °C 400

350

ι

300

.

250

1

1

1

1

145

%J\

55,

50

45

JL 40

35

30

25

20

RETENTION TIME, MINUTES

Figure 1.

Composite chromatogram

matogram at the right defines the front ends i n greater detail, and the one at the left extends the composition 10 carbon numbers higher than the more conventional chromatogram i n the center of the figure. The upperlimit temperature of 400°C was reached near C w i t h this sample. A t this point, only about one third of the amount of sample had emerged, although two thirds of the 50 n-alkanes present were already accounted for. Because n-alkanes are the predominant constituents of ozokerites, determination of carbon number alone identifies most components. T w o adjacent doublets near the beginning of some curves located the two pairs, n-heptadecane-pristane and n-octadecane-phytane, which served as unambiguous counters. F o r samples that started at higher carbon num ber, longer n-alkanes ( C 2 , C 8 , or C ) were added to provide internal standards. In addition, the chromatogram of a heavy paraffinic petroleum 4 0

2

2

3 2


14.

MARSCHNER A N D WINTERS

171

OzokeHteS

containing n-alkanes from 10 to 50 carbon atoms was obtained frequently as an external standard. Amounts of individual n-alkanes were determined by measuring the areas under the successive peaks and above the baseline that represents an unresolved background of other components. Although the background level rises continually with column temperature, the amount relative to n-alkanes increases more from about C o - C o than at either lower or higher carbon numbers. Estimates of the n-alkane content of all samples are included i n Table II. 3


VL-Alhane Abundances

in

4

Ozokerites

Compositions were compared through logarithmic distribution plots of the amounts of successive n-alkanes. Typical distribution curves show pronounced maxima in certain ranges rather than a gradual diminution with carbon number that might have been expected from previous experience with crude oils (12). In many crudes, a maximum occurred at Ce, and the n-alkanes extinguished around C . The maximum at C could have resulted from previous losses of lighter constituents to natural gas or natural gasoline, rather than representing a particularly favored number. The extinction around C could have resulted from previous separation of longer n-alkanes underground or i n tankage, rather than no such material having been present at all. 3 5

6

3 5

In certain ranges of some crude oils, odd-numbered n-alkanes predominate appreciably over the even-numbered homologs. Especially striking are odd predominances from C u to C i that accompany declines i n abundance from C i to C i and from C i s to C o (12). O d d - e v e n predominances were also observed i n ozokerites, but not as prominently and at higher carbon numbers; they w i l l be presented and discussed i n detail elsewhere. Galician ozokerites have n-alkane distribution curves that alomst coincide, as shown by Figure 2. Most abundant are n-alkanes near C 9 , although all three samples have a weak shoulder indicating a somewhat favored range around C i . The Ward's and Warsaw samples are shown with plateaus in the Cso's but the compounds creating them are probably not n-alkanes. Peaks i n the chromatograms i n the C V s of W a r d s sample particularly are doublets if well resolved, and those that appear to correspond to the n-alkanes gradually disappear by C . The second peaks might suggest adulteration, but a natural contaminant seems more probable. Its amount is barely 1%, and it can be found only if the analysis is carried to a high enough carbon number. The curve for the Standard O i l sample has been corrected for the spurious second peaks to better represent the true n-alkane abundances. 9

6

8

2

2

7

5 0


172

SHALE


FUEL

SOURCES


100,

NUMBER OF C A R B O N A T O M S

Figure 2.

Galician ozokerites

Utah ozokerites have a variety of n-alkane distribution curves, as shown i n the three panels of Figure 3. The three ozokerites i n the top panel most nearly resemble those from Galicia: n-alkanes near C 9 are most abundant, and favored ranges are suggested at both lower and higher carbon number. The peaks do not exactly coincide; i t is lowest for Kyune, which has a shoulder at lower carbon number, and is highest for H-23, which has the most prominent plateau at higher carbon number. Unlike those of the Galician ozokerites, the gas chromatogram peaks i n the C V s are not doubled, and the higher constituents are presumably n-alkanes. Fort W o r t h ozokerite i n the middle panel of Figure 3 is displaced to the right of the previous curves, the maximum is broader and asymmetrical, and the lesser peaks are more pronounced. The major peak falls at C 3 3 - C 3 7 , or about four carbon atoms higher than the Galician ozokerites. It has the appearance of an unresolved doublet, with one peak just above C o and the other just above C o. The lower peak falls at C17-C10, no more than two carbons higher than the Galician, whereas the higher peak falls at C59, w h i c h is six or eight carbons higher than either the Galician or the other Utah ozokerites. Finally, i n the bottom panel of Figure 3 are two more Utah ozokerites with four distinct abundance regions. The maximum occurs at C 4 3 - C 4 5 and the next largest peak at C i - C , just opposite of Galician ozokerites and those from Utah in the top panel. The lowest favored region is not w e l l defined, especially for Soldier Summit, and the identification w i t h the C 1 7 - C 1 9 region of the previous curve is uncertain. Indeed, the Wasatch sample may have more than one favored region i n this range. 2

3

4

3

3 3


MARSCHNER

AND WINTERS

Ozokerites


1001

0

10

20

30

40

50

60

70

0

10

20

30

40

50

60

70

0

10

20

30

40

50

60

70

100,

100, LU

NUMBER OF CARBON A T O M S

Figure 3.

Utah ozokerites


174

SHALE


10

20

FUEL

SOURCES

100,


UJ

0

30

40

50

60

70

NUMBER OF C A R B O N A T O M S

Figure 4.

Russian ozokentes

The highest favored region, on the other hand, is well defined at C 5 9 or Ceo and is the same for both samples. The simpler samples i n the top panel came from locations south of the complex ones i n the bottom panel, but the source of the Fort W o r t h sample i n the middle panel is not known. Two other Utah samples were indistinguishable from those illustrated. A n y or all of them may have contained still longer n-alkanes beyond the reach of gas chromatography, but not likely i n large amounts. The Russian ozokerite is compared with the Shor-su ozokerite from southern Fergana, USSR, from the literature (4) i n Figure 4. They are different, but again not unrelated. Shor-su is the simpler, with only two constituents, the smaller at C i e - C i and the larger at C i - C . The other appears to have three constituents, represented b y a maximum at C e, a long plateau at C i 8 - C , and a shoulder near C . These lesser abundances may be the same as the two i n Shor-su. 8

3

3 4

4

2 7

3 5

Distribution curves for other ozokerites that have come to our attention are shown i n Figure 5. Hatchettites from the South Wales coalfield have been thoroughly surveyed ( 7 ) . A single maximum i n n-alkane abundance occurs at C - C , with 12 of 18 samples actually peaking at C , but the chromatograms d i d not extend beyond C , and longer n alkanes could have been missed. Before weathering, another hatchettite showed a shoulder near C i that suggests a lesser abundance at lower carbon number. E i n H u m a r ozokerite from the eastern side of the Dead Sea i n Jordan (13), shows an abundance maximum at C , together with the suggestion of a lower abundance anomaly. 2 4

2 9

2 7

3 5

6

3 8


14.


175

Ozokerites

Patterns seen in the Russian ozokerites of Figure 4 and the miscellaneous ozokerites of Figure 5 were previously visible in the Utah ozokerites of Figure 3. The most abundant n-alkanes may fall either near C , as i n Galician ozokerites, or near C , as i n the Soldier Summit type, and implications based on samples from Galicia alone could be misleading. 2 9

4 e


Causes of Maxima in Abundance A single maximum in abundance is deceptively easy to explain. A diminution with increase in carbon number might be expected from the thermodynamics of whatever chain-forming process was originally i n volved, a diminution with decrease i n carbon number might be expected from such separation processes as volatility or solubility that might have occurred subsequently, and the resultant of the two processes would be an intermediate maximum, whether the material under investigation was an ozokerite or a crude oil. One difficulty with this ready explanation is that the maxima i n ozokerite and petroleum are not the same. Three of eight crude oils peaked at C u and three more at C or below (12). Two other crude oils peaked higher, as shown in Figure 6 (State L i n e at C i and Uinta Basin at C 2 7 ) , but only Uinta Basin has a maximum at a carbon number nearly as high as those i n ozokerites. Inspection of Figure 6, however, suggests the possibility of coincidence; Uinta Basin has a distinct plateau near C and an indistinct one around C19, whereas the broad peak for State L i n e could result from overlapping of three abundance maxima at C , C i , and C . Closer examination of n-alkane abundances among ozokerites support this possibility. A l l three Galician samples, for example, show an abun7

9

9

9

2 9

100 •

UJ

1

0

1

K)

20

1

30

40

50

60

1

70

NUMBER OF CARBON A T O M S

Figure 5.

Other ozokerites


9

176

•RCENT (LOGARITHMIC SCALE)

SHALE


CL

O I L , T A R SANDS, A N D R E L A T E D

FUEL

SOURCES

100

10.0 •

STATE

ftjX^>n UINTA BASIN

1.00

~i 0.10

1

0.01 0

10

20

30

40

50

60

70

NUMBER OF CARBON ATOMS

Figure 6.

Crude oils

dance shoulder around C i that could correspond to the maximum i n State L i n e and the indistinct shoulder i n Uinta Basin. Russian ozokerite and Pencoed hatchettite are similar. The lowest abundance anomalies for Utah ozokerites occur only slightly lower, as also does that for unweathered Blaenavon hatchettite. 8

Multiple maxima such as occur with Russian and some of the Utah ozokerites are more difficult to explain. Nearly a l l ozokerites show a pronounced abundance peak near C i , but several show a clear secondary peak around C e, and a few show even a third one in the C s . Variation also remains to be accounted for. The prime maxima for hatchettites occur four or more carbon atoms lower, and the secondary peak for E i n Humar occurs perhaps eight carbon atoms lower. 3

4

50

The presence of more than one favored abundance range suggests that several processes were involved in ozokerite formation. If the processes occurred i n sequence, abundances representing vestiges of earliest processes as well as n-alkanes formed latest might both be seen. T o account for the distribution of n-alkanes actually found i n petroleums and ozokerites would require at least the following processes. 1. Vaporization. Maxima i n n-alkane abundance ranging from C , to C u i n most crude oils probably result from the vaporization of normally gaseous constituents, primarily through reduction i n pressure. The position of the maximum should depend primarily on the gas-to-oil ratio, greater ratios giving higher maxima. r

2. Biosynthesis. The favored C15-C19 range may represent vestigal odd-numbered n-alkanes produced by decarboxylation of the even-numbered 16-20 carbon fatty acids produced by plants. Presumably the mechanism for the biosynthesis is the same for ozokerite as for pertroleum,



14.


177

Ozokerites

but any differences in the positions of the maxima would result mainly from differences i n the source plants. 3. Maturation. The low level of odd-even predominance categorizes ozokerites as having undergone maturation in a manner similar to equilibrium crude oils. The light gases involved i n vaporization (1.) would also be produced in the maturation process. 4. Evaporation. Maxima i n the approximate range C20-C30 are partly the result of evaporation of lower homologs during long access to the atmosphere. This process is inherent in that of "inspissation," through which deposits of bitumen are conventionally accounted for, and has been implicated i n hatchettite formation (7). 5. Melting. Pronounced maxima in the C 2 7 - C 3 3 range may result in part from the loss of smaller n-alkanes by a fractional melting process analogous to "sweating" i n wax manufacture. Positions of the maxima would be affected by overlap with evaporation (4.). 6. Solubility. Also involved i n concentration of n-alkanes i n the C - C 3 range may be the immobility of higher n-alkanes by reason of insolubility in mixed hydrocarbons underground. Immobility may be assisted by adsorptive forces involving minerals or such organic materials as kerogen. 7. Selective synthesis. Abundant n-alkanes i n the C 8 - C o range must obviously have been formed by selective synthesis at some stage of the genesis—early, by plants; during diagenesis, perhaps accompanying maturation; or subsequently, by some unfamiliar polymerization process. Perhaps the traces of nonnormal structures in Galician ozokerites provide a clue to the mechanism involved. 8. Inertness. The final favored range of n-alkane abundance at C 5 1 - C 5 9 may represent vestiges of an infinite series that survived by reason of inertness through, say, insolubility. O n occasion, n-alkanes above C o were observed, and homologs as high as C100 may be present in minute amounts. Not all these processes need necessarily have been involved in formation of every petroleum or even every ozokerite, but both seem to have required a considerable combination of circumstances, with ozokerite the more complicated of the two. The variations in its composition, as well as its scarcity relative to petroleum, coal, and lignite, were inevitable results. 2 7

3

3

5

8

Acknowledgments W e thank Arie Nissenbaum for making the chromatogram for E i n Humar ozokerite available in advance of publication. Literature

Cited

1. Anders, Edward, Hayatsu, Ryoichi, Studier, Martin H., "Organic Compounds in Meteorites," Science (1973) 182, 781.


178

SHALE OIL, TAR SANDS, AND RELATED FUEL SOURCES

2. Redwood, Boverton, "Galician Petroleum and Ozokerite Industries," J. Soc. Chem. Ind. (1892) 11, 93. 3. Kwiecinska, Barbara, Ratajezak, T., "Ozokerite from Starunya," Bull. Acad. Pol. Sci. (1969) 17, 155, Chem. Abs. (1970) 73, 132948. 4. Kovjazin, V. E., Hala, S., "Presence of C -C Isoprenoid Alkanes in Ozokerite," Collect Czech. Chem. Commun. (1973) 38, 2938. 5. Robinson, Heath M., "Ozokerite in Central Utah," U.S. Geol. Surv. Bull. (1916) 641A. 6. Abraham, Herbert, "Asphalts and Allied Substances," Vol. I, Chapter7, Van Nostrand, 1960. 7. Firth, J. N. M., Eglinton, G., "Hatchettite from the South Wales Coalfield," Adv. Org. Geochem. (1971) 613. 8. Louis, M., Bienner, F., "A Specific Paraffinicity Index: Hatchettite," Rev. Inst. Fr. Pet. (1954) 9, 149. 9. Kastner, Dietrich, Moos, Josef, Schultze, Georg R., "Composition of Crude Polish Ozokerite," Erdoel Kohle (1959) 12, 77. 10. Ivanovsky, Leo, "Ozokerite and Related Substances," 3 vols., Hartleben, Leipzig, 1934. 11. Rozanova, E. D., Shturm, L. D., "Chances in Ozokerite Composition under the Action of Microorganisms," Mikrobiologiya (1966) 35, 138. 12. Martin, Ronald L., Winters, John C., Williams, Jack Α., "Distribution of n-Paraffins in Crude Oils and Implications to Origin of Petroleum," Nature (1963) 199, 110. 13. Nissenbaum, Α., Aizenshtat, Z., "Geochemical Studies on Ozokerite from the Dead Sea Area," Chem. Geol. (1975) 16 (2), 121.


15

20

RECEIVED December 16, 1974.


Shale Oil, Tar Sands, and Related Fuel Sources - ACS Publications

Recommend Documents